Shedding New Light

8/13/2019 Shedding New Light

1/6

featureartitll

Shedding new light on light in the oceanTommy D. Dickey, George W. Kattawar, and Kenneth J. Voss

Recent advances are making it possible for optical oceanographers to solvea host of pressing environmental problems.

Tommy Dickey holds the Secretary of the Navy/Chief of Naval Operations Chair in Oceanographic Sciences at the University of California,Santa Barbara. George Kattawar is a professor in the physics and astronomy department and the Institute for Quantum Science and Engi-neering at Texas A&M University, College Station. Kenneth Voss is a professor and chair of the physics department at the University ofMiami in Florida.

Light drives the physics, chemistry, and biology of theocean. Without light in the ocean, life on Earth could not exist,nor would there be fuel for the heat engine that drives theocean's currents and the atmosphere's circulation. Further-more, light and sound are the two primary means availablefor probing the ocean. (See the article by Tom Sanford, KathieKelly, and David Farmer in PHYSICS T o d a y , February 2011,page 24.)

Optical oceanography, which concerns all facets of light,its interactions with seawater, and its ultimate fate, is centralto many important studies.12It is vital for addressing prob-lems of photosynthesis, ecosystem dynamics, the health ofthe ocean, seawater clarity, underwater imaging, biogeo-chemical cycling, carbon budgets, upper ocean thermody-namics, and climate change. The future of the ocean and itsinhabitants will depend on our ability to learn how anthro-pogenic activity affects them.

Light interactionHow does light interact with seawater? The simplest opticalinteractions, which occur at the boundary of the atmosphereand ocean, are governed by Snell's law and the Fresnel equa-tions. But even those straightforward geometrical optics ef-fects are complicated by surface roughness. Once light entersthe ocean, it interacts with water molecules and with agreat variety of other constituents, organic and inorganic.Accurately modeling those complex interactions requiresradiative transfer methods and accurate measurement of themany highly variable constituents and parameters in thewater. It would be convenient to think of the ocean as simplya homogeneous body of water. But the ocean is in fact more

like a stratified witches' brew, with many different solid anddissolved ingredients.The interactions of light in the ocean involve many

processes and feedback mechanisms. Consider phytoplank-ton, microscopic drifting algae that form the base of the oceanfood chain by performing photosynthesis. They absorb light,take up C 02, and release oxygen. They also fluoresce some ofthe light back into the ocean, and they can exhibit biolumi-nescencelight emission through diemical processes.

Phytoplankton vary in size, shape, and optical proper-ties, and they contribute to the color of seawater. They affectthe heating and stratification of the upper ocean throughtheir interactions with light. And those processes in turn af-

fect their growth environment. Anand Gnanadesikan andcoworkers at Princeton University have recently suggestedthat phytoplankton may even affect the frequency of ty-phoons.3It had been suggested in 2004 that hurricanes, whichincrease the availability of plant nutrients in the upper layer,can stimulate phytoplankton blooms and thus accelerate thedrawdown of C02by the ocean.4

Why is the ocean blue?Speculation about the blue color of the ocean, as seen fromabove, goes way back. Lord Rayleigh claimed it was simply re-flection of the blue sky. The correct explanation required com

H2wI


2/6

birring the 19th century ideas of Robert Bunsen, who felt thatthe color depended on light absorption by water, and JacquesLouis Soret, who felt that the color was entirely due to scatter-ing. C. V. Raman pointed out the importance of molecular scat-tering, and in 1923 Vasily Shuleikin combined those ideas todevelop a complete explanation of the color of the sea.

In very clear, open waters with few particulates, calledhydrosols, the ocean's optical properties depend primarily onthe scattering and absorption properties of the water moleculeitself. Let us first address absorption. One might think that thewavelength dependence of water's absorption coefficientasine qua non for calculating the ocean's intrinsic color as afunction of depthis known to high precision. But one wouldbe wrong. Figure 1 shows the recent history of the absorptioncoefficient's measurement in pure water. The principal featureis the almost thousandfold rise from the absorption minimumat blue wavelengths to a maximum in the red.

The very little absorption on a scale of meters in the blueis why there have been such large discrepancies in the meas-urements at that end of the visible spectrum. The minimal ab-sorption arises primarily from the vibration modes of thewater molecule's O H bonds. The fundamental vibrationalmodes occur at IR wavelengths near 3 jim. Therefore, absorp-tion in the blue requires excitation of higher lying states vi-brating at high overtones of the fundamental. Because thedensity of such states is low, there's little absorption at bluewavelengths. It's remarkable that the intrinsic blue color ofwater is due primarily to molecular vibrations and not to di-rect electron interactions, which are the primary determi-nants of color in almost all other substances.

Scattering in waterThe scattering of light in the ocean can be elastic (with no fre-quency shift of the incident radiation) or inelastic (frequencyshifted). Both types are often called Rayleigh scattering. ButRayleigh certainly did not describe inelastic scattering in bulkliquids; that phenomenon was not understood beforeRayleigh's death in 1919. (See the article by Andrew Youngin P h y s ic s T o d a y , January 1982, page 42.)

The scattering of light in liquid water is primarily due todensity fluctuations, which can be broken down into two dif-ferent types, mechanical and thermal (see figure 2a). Mechan-ical disturbances create isentropic pressure fluctuations thatpropagate as phonons with the speed of sound in water. Lightof incident frequency a)0scattered off such a phonon suffers apair of frequency shifts to produce a so called MandelstamBrillouin (MB) doublet, as shown in figure 2b. In the languageof Raman scattering theory, the doublet's red and blueshiftedspectral lines are called, respectively, the Stokes and antiStokes lines. The weaker unshifted line in the spectrum ofscattered light, discovered by Evgenii Gross in 1930, is due toscattering off nonpropagating, isobaric entropy fluctuations.5

Bragg scattering off phonon waves and Doppler shifts

due to the motion of those waves, as shown in figure 2b, de-termine the MB frequency shifts a = o)0

wmbH = 2n sin(6/2)v l c , (1)where n, the refractive index of water, depends on ajg, salinity,and temperature; 6 is the scattering angle at which the scat-tered waves are observed; and vs and c are, respectively, thespeed of sound in the water and the speed of light in vacuum.

Note that there is no frequency shift in the forward di-rection, and that the maximum shift occurs at 180 backscattering. For pure water, the backscattering frequency shift isaround 7.5 GHz, which is also the frequency of the soundwaves producing the light scattering.

Light scattering in water

Mechanical disturbances1 I

Thermal processesl~ l

Density or refractive index fluctuationsI I

Pressure fluctuations(phonons)

1 iSound waves

[ IMB doublet

H ib

I IEntropy fluctuations

1 1Diffuse thermal dissipation

I IGross line

Figure 2. Light scattering in water by fluctuations in densityor refractive index, (a) Mechanical disturbances generatephono ns that scatter incident light of frequency at0 to producea Mandel stam B rillouin (MB) doublet of spectral lines shiftedredward and blueward by wMB, give n by equation 1 in the text.Thermally generated entropy fluctuations, by contrast, scatterlight without shifting its frequency, resulting in the Grossline in the spectrum of scattered light, (b) One can deriveequation 1 by imagining the Bragg scattering of incidentphotons of momentum k off phon on waves (shown asstriations) of incident momen tum q . Scattering off approxi-mately approaching (+q) or receding (-q) wavefronts Dopplershifts the photon, respectively, toward the blue or red. Thewavefront orientation of the relevant phon ons is deduce dfrom the scattering angle 9 at which the observer is looking.

The ratio of the intensity of the unshifted Gross line tothe sum of the intensities of the MB doublet lines in the reflected light spectrum is (CP/Cv) 1 , where CPand Cv are, re-spectively, the specific heats of water at constant pressure andconstant volume. Because that ratio at 25 C is only about0.01, the spectrum of scattered light in pure water shows verylittle intensity at the unshifted wavelength; almost all thescattered light appears in the Stokes and anti Stokes lines.

www.physicstoday.org April 2011 Physics Today 45
http://www.physicstoday.org/http://www.physicstoday.org/


3/6

Figure 3. (a) The Multi-Angle Scattering and Optical Transmission (MASCOT) detector built and deployed by Michael Twardowski measures light absorption and light scattering properties and bubble size distributions of ocean water at different loca-tions and depths. (Photo by Michael Twardowski.) (b) Optical system used by a group from the Scripps Institution of Oceanog-raphy for measuring rapid light fluctuations that occur naturally just below the ocean's surface. (Photo by Dariusz Stramski.)

That result has enormous consequences for improving imagequality in the ocean, because light scattered from hydrosolsor underwater objects experiences no frequency shift.

Another inelastic process in water is Raman scattering,where vibrations of the water molecule modulate the liquid'spolarizability and thus give rise to both Stokes and antiStokes Raman shifts of title incident light. There's always acompetition between Raman scattering off single water mol-ecules and off polymeric stacks of water molecules held to-gether by hydrogen bonds. The resulting equilibrium is verytemperature sensitive, and it can alter the shape of the Ramanband. That effect has been exploited for measuring depthprofiles of water temperature with light detection andranging (lidar) systems.6

Although the Raman scattering coefficient is about thesame as the MB scattering coefficient, there are important dif-ferences between the two processes. Used together, they pro-vide a powerful tool for remotely monitoring the soundspeed, temperature, and salinity in the ocean as a function ofdepth. Like Rayleigh scattering in the atmosphere, bothprocesses have a roughly A4 dependence on wavelength,which contributes to the ocean's blue color, as it does to theblue sky. With minimum absorption near 420 run, in somevery dear, dean locations, most notably near Easter Island inthe South Patif ic Gyre, the water appears almost purple.7

It's useful to define two dasses of water's optical prop-erties: Inherent optical properties (IOFs), which characterizespectral scattering and absorption independent of the direc-tional structure of the ambient light field, can be measured inthe lab. Apparent optical properties (AOPs), on the otherhand, depend on the light Held and can be measured only insitu. Radiometric quantities include the radiance L, definedas the light power impinging on an underwater detector perunit area, frequency, and acceptance solid angle. It depends,of course, on the detector's underwater position and the di-rection it's pointing, and on the Sun's position, (Irradiance isthe integral of radiance over some specified solid angleforexample, the entire downward hemisphere.) To predict radi-ance and irradiance for comparison with measurements oneuses radiative transfer models that take account of all thethings that can happen to a photon on the way from the Sunto the detector, using the lOPs as inputs.

46 April 2011 Physics Today

Radiance and irradiance thus computed and comparedwith measurements are often used for determining AOPs. Forexample, the spectral diffuse attenuation coefficient at a givenoptical wavelength is defined as the negative of the verticalgradient of downward irradiance.1 In natural ocean water,many factors modify optical properties, and hence the appar-ent color. They indude dissolved organic matter, which pref-erentially absorbs in the UV, thus shifting the water's color to-ward green and red. That shift is often very evident in puddlesor areas of standing water in which leaves are decaying.

In areas of sediment runoff, large partides dominate thescattering, decreasing the wavelength dependence of thebackscattering. When that effect combines with dissolved or-ganic material, the water acquires a brownish tinge. Phyto-plankton contain chlorophyll, which absorbs light in the blueand red. Thus an increase in phytoplankton concentrationcauses the ratio of blue to green reflection from the ocean todecrease, creating greenish hues. In shallow water, reflectionfrom the sea floor can also have a profound effect on the sea'sapparent color. That effect can be exploited in shallow watersfor bathymetrydetermining sea floor depth.

Observational methods and shafegiesOptical oceanography relies heavily on field data setsneeded for discovering new phenomena and developingmodels. Some of the earliest ocean measurements of lightwere done with a Secchi disk, named after the 19th centurypriest and astronomer Pietro Angelo Secchi, who also didoceanographic studies aboard the papal yacht L'Immacolata Concezione. The white disk, about the size of a dinner plate,is lowered slowly in the sea until, at the so called Secchidepth, it can no longer be seen from above the surface. Sosimple a measurement turns out to be amazingly stable fromobserver to observer, and it can be linked to modem globalscale optical and biological data.8

One objective of Britain's 1872 76 HMS Challenger expedi-tion, which essentially initiated oceanography as a science,was to carry out worldwide measurements of light penetrationinto the sea. Soon thereafter, color based comparators inventedby Swiss investigator F r an^ois Alphonse Forel were used todefine water masses by their color. Toward the end of WorldWar E, electronic radiometers began being used for oceanic op-

www.physicstoday.org


4/6

tical measurements. Those instruments, like the present gen-eration of optical instruments, had to withstand shock duringdeployment and recovery, high pressure, temperature varia-tions, corrosion, and fouling by marine organisms.

Today's optical instruments are being used to investigatehow light interacts with seawater for a variety of purposes:to determine IOPs and AOPs, particle size distributions,chemical concentrations, and resident organisms.2 To suchends, the instruments measure the absorption, scattering,and total attenuation of light. They utilize either ambient sun-light or artificial light to quantify the optical medium and thelight field itself.

In the past decade, newly developed ocean instrumentsusing linear variable filters and spectrometers have increasedspectral resolution by almost an order of magnitudeto afew nanometers in the visible. This so called hyperspectralresolution makes possible improved identification and quan-tification of assemblages of phytoplankton and dissolvedmatter. It also provides detailed bathymetric data critical foraddressing a host of practical and research problems.9

RaDyOA fundamental IOP is the volume scattering function, whichdescribes the angular scattering pattern of light. Such meas-urements are difficult because they span a large dynamicrange; variations with angle span several orders of magni-tude. But the past few years have seen much progress. As oneexample, during the 2008 09 Radiance in a Dynamic Ocean(RaDyO, pronounced "radio") program conducted off Cali-fornia and Hawaii, oceanographer Michael Twardowski(WET Labs, Narragansett, Rhode Island) and colleagues de-ployed a newly developed instrument package, shown in fig-ure 3a, capable of absorption measurements at several wave-lengths and scattering measurements at 17 different angles.Data from such instruments provide essential inputs for radiative transfer models. Twardowski's instrument has alsobeen used to measure bubble size distributions (rangingfrom submicron to a few hundred microns), complementingbubble measurements done with acoustic resonator systems.

Until recently, few studies have been devoted to the na-ture and potential effects of high frequency light fluctuationswithin 10 meters of the ocean's surface. Such fluctuationsarise from intense sunlight focused and defocused by waves.As part of the RaDyO experiment, Dariusz Stramski andcoworkers at the Scripps Institution of Oceanography in La

Jolla, California, deployed an instrument suite comprisingseveral irradiance and radiance sensors that can measurelight fluctuations on time scales from milliseconds to minutes

(see figure 3b). Stramski and company showed that instanta-neous light pulses exceeding 10 times the average irradianceoccur near the surface. The implications of such bursts forheating and photosynthesis remain to be explored.

Virtually all of the IOPs and light field characteristicscan be derived from the radiance distributionthat is, themeasurements of radiance at all angles.1During the RaDyOprogram, advances were made in the accomplishment ofsuch demanding measurements by Marlon Lewis (DalhousieUniversity, Halifax, Nova Scotia), one of us (Voss), andcoworkers. Lewis's ship deployed instrument measured theradiance distribution as a function of depth and time with anangular resolution of less than 1 over a dynamic range of amillion. Cameras were also developed for measuring polar-ized spectral radiance distributions in the water and abovethe surface at five different wavelengths. Voss and coworkersfound that near the surface under dear skies, the polarizationof the light field was dominated by refracted skylight. Atdepths below 20 m, the polarization was due mostly to lightscattering by the water itself.

Polarization is enormously important to many sea crea-tures. Some use polarized light as a compass, much in the sameway honeybees do. Squid use polarized light to see the phyto-plankton on which they feed. It's also conjectured to be impor-tant for the ability of squid and other cephalopods (octopusand cuttlefish) to camouflage themselves.10(See figure 4.)

Modeling of the surface and subsurface light field re-quires inputs of high spatial and temporal resolution. Oneneeds to know the topography of the surface and its winddependent conditionroughness, breaking waves, whitecaps, or foam. Optical measurements of waves down to mil-limeter scales were carried out by several groups duringRaDyO. They used lidar, IR, and polarimetric video cam-erassome of them stereographic. Chemical and physicaldata were also collected in the surface microlayer down to1 mm. The light field in that microlayer affects surface tensionand photochemical transformations.11

Calculating radiative transferTo model complex, rapidly time varying features in theocean, one must confront the time dependent radiativetransfer (RT) equation that seeks to describe energy transferby electromagnetic radiation in media. Models have to cou-ple the atmosphere to the ocean across a time varying inter-face consisting of waves whose structural detail ranges frommillimeters to meters. And the waves can be nonlinear.

Many methods are available for solving the static one-dimensional RT equation for either the atmosphere or the

t = 0 270 ms 2.07 sFigure 4. An octopus of the common species Octopus vulgaris reacts to the approach of a diver by completely changing itscamouflage within two seconds. Initially camouflaged to match the mottled greenish color, pattern, texture, and reflectiveintensity of the surrounding algae, it quickly changes to match the white sea floor. (Adapted from ref. 10.)



5/6

Figure S. R/P FLIP, a floating research platform operated by the Scripps Institution ofOceanography, is an extraordinarily stable platform for the dep loyme nt of multiple researchinstruments at sea. It's a towable sea going vessel 108 meters long that b ecome s the tall, sta-ble tower shown in the p hoto w hen its aft ballast compartments are flooded to flip the vesselbow up. (Photo by Kyla Buckingham.)

ocean.12 But the number of methods becomes quite smallwhen total coupling is requited. Modelers are faced withtime varying patterns requiring resolution down to a thou-sandth of a second. That's in addition to the range of relevantspatial structures. So 3D, time dependent models are manda-tory. Recently, a group at Texas A&M University led by one

of us (Kattawar) and an MTT group led by Richard Yue haveindependently developed 3D Monte Carlo simulation mod-els for treating that daunting problem. Many of the RaDyOmeasurements include polarization data. So the models hadto handle the vector RT equation; solving the scalar RT equa-tions almost always gives incorrect results, with the magni-tude of the error depending strongly on the degree of polar-ization of the light field.

The IOPs needed for performing the requisite RT calcu-lations are the volume scattering function (involving a ma-trix if polarization is taken into account), the single scatter-ing probability, and the total beam extinction coefficient.The computational methods developed by the MIT andTexas A&M groups have yielded very good agreement withmany of the RaDyO measurements.13Linear dimensions in

RT calculations are usually given in terms of a dimensionless"optical depth,"the actual length divided by the charac-teristic extinction length for light in a given environment.

Remote color sensingSatellite based remote observation of ocean color has mademajor contributions to oceanography. The variation of colorwith chlorophyll concentration provides remote sensingtechniques for estimating biomass. The first ocean color satel-lite instrument, the experimental Coastal Zone Color Scan-ner, was flown aboard a National Oceanic and AtmosphericAdministration satellite in 1979. With pioneering data fromthe CZCS, algorithms were developed for turning measured

radiance into the relevant bio-geophysical quantities.14 TheCZCS experiment was remark-ably successful, and the instru-ment continued to make meas-urements aboard the satellite foreight years. Its name notwith-standing, the CZCS was mostuseful in the open ocean, where

color is determined by fewer fac-tors than in coastal zones. In theopen ocean, chlorophyll concen-tration is the primary determi-nant.

Coastal zones, complicatedby sediment runoff, dissolvedorganic matter, and shallowwater effects, require more spec-tral information than could beprovided by the four visiblelight bands of the CZCS. Latergenerations of color sensors withadvanced spectral capabilities,flown by US, European, Indian,

and Japanese space agencies,have advanced the remote mon-itoring of the ocean. Applica-tions range from local issuessuch as fisheries and harmfulalgal blooms in the coastal re-gions to dimate issues and thecycling of carbon in the globalocean environment.2

Oceanographers have to measure many variables in acomplex, inhomogeneous medium in an ever changing envi-ronment. To that end they have to deploy their instrumentsaboard a dazzling variety of specialized platforms, rangingfrom orbiting satellites and airplanes to twin hulled catama-rans, undersea vehicles, and R/P (for research platform) FLIP,

a very stable floating instrument platform that comes intotemporary existence when the 108 meter long vessel isflipped over so that its bow rises straight up, as shown in fig-ure 5. A video by University of Rhode Island oceanographerHelen Czerski that illustrates several ocean optical measure-ments made from various platforms during the RaDyO pro-gram is available on YouTube.15

Visibility and imagingThe study of visibility in the ocean, and the associated prob-lem of seeing underwater objects, was one of the earliest ap-plications of ocean optics. Water is, of course, denser than air,and it generally harbors more particulates. In the ocean, max-imum imaging distances are highly variable, ranging fromcentimeters in dirty water to more than 100 m in very dearwater.

In the ocean, scattering by particulates and by the wateritself affects image quality in many ways. A quantitative meas-ure of the visibility of an object is the contrast, C, given by

C(r) {Lc(r) - L h(r))/Lh(r)/ (2)

where r is the distance of the object from the observer, and L0and are the radiances of the object and its background.With that definition, we have the "inherent contrast" at r = 0,the "apparent contrast" at r, and the "contrast transmit-tance," whidi propagates the inherent contrast to the appar-ent contrast.

48 April 2011 Physics Today www.physicstoday.org


6/6

Much of the theory and experimental work on imagetransmission in the ocean was developed by the Scripps Vis-ibility Laboratory in La Jolla, California, starting in the1940s.16In general, the contrast of a relatively large object isreduced by scattering of ambient light in the medium be-tween the viewer and the object, causing what's known asveiling, or path radiance. The ambient light increases equallythe apparent radiance from the object and the background,thus effectively reducing the object's apparent contrast. Forlarge objects, that's often the limiting factor for visibility.

For longer distances or smaller objects, small angle for-ward scattering of light reflected from the object can causeimages to blur, which reduces the contrast between light anddark portions of the object and, along its edges, between ob-

ject and background. In water, small angle scattering is alarge portion of the total scattering of light by particles. Theresultant blurring can be quantified by means of the pointspread function (PSF) often used to characterize the limits ofimaging devices such as telescopes. In the early 1970s,Willard Wells at Tetra Tech Inc in Pasadena, California, pio-neered the use of PSFs to enhance underwater imaging, andsuch work has continued to yield new techniques for betterimaging. Related research has sought to understand the dif-ferent methods by which aquatic animals camouflage them-selves against predation (see figure 4).

Methods intended to improve imaging in the water workeither to reduce the effects of the veiling backscattered light orto reduce the blurring effect of forward scattering. One can, forexample, illuminate the underwater object with a polarizedlight source, cover the camera's lens with a polaroid filter, andthen time gate both source and camera so as to minimize therecording of light not directly reflected off the object.17

To compensate for the blurring effect of forward scatter-ing, imagers have recently revisited time varying intensity(TVI), an approach first used at the Scripps Visibility Lab. Thetechnique, currently being investigated by groups at theNaval Air Systems Command and elsewhere, uses a laserline scan illumination source dose to the object. Rather thanbeing a conventional multipixel imaging device, the TVI im-ager is a single pixel detector synchronized to the scanningof the source. In that way, forward scattering between the ob-

ject and the detector does not degrade the image. (The limit-ing factor is blurring between the illumination source and theobject.) Recent work on TVI systems has demonstrated thepossibility of underwater imaging at optical depths of 25that is, 25 times the distance in which the medium attenuatesthe direct radiance by a factor of e.

With regard to the purposeful avoidance of imaging,groups at the Marine Biological Laboratory (Woods Hole,Massachusetts) and elsewhere are studying the ability ofcephalopods and some fish to camouflage themselves. Asshown in figure 4, such a creature can, in response to achanged background, very quickly change its skin's spectral

reflectance, patterning, texture, and even the polarization ofthe light reflected from its skin. Current work strives to un-derstand the structures in the skin that can do all that, andhow the animal determines the best camouflage strategy.

Toward the futureWith relatively modest spectral resolution, researchers havebegun to quantify near surface constituents in the ocean suchas chlorophyll, dissolved organic and detrital materials, andeven particle size distributions. Data from instruments withsignificantly higher spectral resolution, deployed in situ orcarried aboard space and aircraft, are now yielding im-proved mapping of shallow ocean depths and the plants

growing there and are even distinguishing between classesof phytoplankton. New multidirectional measurements oflight on short time and space scales will allow oceanogra-phers to better characterize the ambient light and its inter-actions in the ocean.

Fluorescence techniques are making it possible to extractdetailed information about phytoplankton and oil spills. Ac-tive sensing with lasers has great potential for revealing in-formation about the physiological states of phytoplankton.Pulsed narrowband lasers will let us exploit MB and Ramanscattering to probe vast regions of the ocean for salinity, tem-perature, and sound speed as a function of depth. Femtosec-ond laser pulses with intensities high enough to affect mate-rial properties of the medium may someday be used forhyperspectral remote sensing. Adaptive optics, which as-tronomers use so effectively to combat atmospheric distor-tion, may well be applicable for reducing the effect of wavesand turbulence on imaging in the ocean. Much new informa-tion can also be garnered by exploiting polarized light.

The next generation of optical oceanographers has muchto accomplish. The potential number of ocean variables thatcould be measured optically is great. Multidisciplinary, mul-tiplatform sampling of the ocean has already demonstrated

its effectiveness. Still, no truly concurrent measurementshave been accomplished over the space and time scales rele-vant to most problems. Future oceanographers will be ableto capitalize on multiscale observing systems and on power-ful computer systems that can handle integrative dataassimilation models. But they will need to be well versed intraditional subdisciplines, with ocean optics at the nexus, aswere the first oceanographers aboard HMS Challenger a cen-tury and a half ago.

We acknowledge support by the Office of Naval Research's Ocean Optics Program and thank our collaborators in the RaDyO program and its manager, Steve Ackleson, for their input and support.

References

1. C. D. Mobley, Light and Water: Radiative Transfer in Natural Waters, Elsevier, New York (1994).2. T. Dickey, M. Lewis, G. Chang, Rev. Geophys. 44, RG1001 (2006),

doi:10.1029/2003RG000148.3. A. Gnanadesikan et al., Geophys. Res. Lett. 37, L18802 (2010), doi:

10.1029/2010GL044514.4. J. M. Babin, J. A. Carton, T. D. Dickey, J. D. Wiggert, ]. Geophys.

Res. 109, C03043 (2004), doi:10.1029/2003JC001938.5. P. A. Fleury, H. P. Boon, Adv. Chem. Phys. 24,1 (1974).6. D. A. Leonard, B. Caputo, F. E. Hoge, Appl. Opt. 18,1732 (1979).7. A. Morel, J. Claustre, B. Gentili, Biogeosciences 7, 3139 (2010).8. D. G. Boyce, M. R. Lewis, B. Worm, Nature 466, 591 (2010).9. T. Dickey, Oceanography 17(2), 9 (2004).

10. R. Hanlon, Curr. Bio. 17, R400 (2007).11. O. Wurl, L. Miller, R. Rottgers, S. Vagle, Mar. Chem. 115,1 (2009).12. G. W. Kattawar, ed., Selected Papers on Multiple Scattering in Plane

Parallel Atmospheres and Oceans: Methods, SPIE Milestone Series,vol. MS42, SPIE Publications, Bellingham, WA (1991).

13. Y. You, D. Stramski, M. Darecki, G. W. Kattawar, Appl. Opt. 49,1041 (2010).

14. H. Gordon, A. Morel, Remote Assessment of Ocean Color for Inter pretation of Satellite Visible Imagery: A Review, Springer, New York(1983).

15. RaDyo videos are available at http://www.youtube .com/watch?v=vym7TQBc TE, http://www.youtube.com/ watch?v=B Zsrj3Lg5Q, and http://www.youtube.com/watch7v=01np lgp3jg.

16. S. Q. Duntley, ]. Opt, Soc. Am. 53(2), 214 (1963).17. J. S. Jaffe, K. D. Moore, J. McLean, M. P. Strand, Oceanography

14(3), 64 (2001).18. R. M. Pope, E. S. Fry, Appl. Opt. 36, 8710 (1997).

http://www.youtube/http://www.youtube.com/http://www.youtube.com/watch7vhttp://www.physicstoday.org/http://www.physicstoday.org/http://www.youtube.com/watch7vhttp://www.youtube.com/http://www.youtube/

Shedding New Light

Documents

Transcript of Shedding New Light